INDEX

Explanations

references to specific past US presidential administrations, focusing particularly on the Obama administration

oai_token-act-pair · gpt-3.5-turbo

references to the Obama and Trump administrations

oai_token-act-pair · gpt-4o-mini Triggered by @bot

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 9-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.9.hook_resid_pre

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

Skylion007/openwebtext

Features

24,576

Data Type

torch.float32

Hook Point

blocks.9.hook_resid_pre

Architecture

standard

Context Size

128

Dataset

Skylion007/openwebtext

Hook Point Layer

Activation Function

relu

Embeds

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

NetMessage

-0.85

aterasu

-0.75

...]

-0.72

 Takeru

-0.68

 Lydia

-0.65

 Korra

-0.64

asma

-0.62

hemy

-0.62

Leod

-0.61

 sqor

-0.61

POSITIVE LOGITS

 administration

1.39

 presidency

1.22

 Administration

1.21

 campaign

1.08

 agenda

1.05

era

1.03

 playbook

1.00

 Presidency

0.98

 administrations

0.95

 regime

0.92

Activations Density 0.093%

references to specific past US presidential administrations, focusing particularly on the Obama administration

references to the Obama and Trump administrations

No Comments

No Known Activations